| Summary: | Background In 2013, following reminders about statutory notification of HIV/AIDS, Portuguese clinicians notified all cases, irrespective of previous notification. At the same time, hospitals introduced electronic records that automatically generated additional paper notification reports. No supplemental resources were available to process the 10-fold increase in notifications. The ensuing backlog caused delays in providing timely information for HIV program planning and evaluation. We investigated whether management principles from the automobile industry (LEAN) could improve data management efficiency. Methods Efficiency was defined as the time spent processing the paper report into electronic surveillance information. We used value stream mapping to understand the process and employed focus groups to identify areas for improvement (LEAN methodology). We recorded the time taken to complete this process for randomly selected batches of reports and calculated the average time per report. Results When consulted, stakeholders expressed the need for information about recent HIV/AIDS diagnoses. We prioritized processing cases diagnosed between 2011-2013. We reduced data-entry errors and transcribing time by inserting drop-down menus and automatic variable calculators. We implemented auto-search during data entry to prevent duplication. We redesigned the data entry mask to match the paper report. Before intervention, processing time was 9 minutes and 28 seconds (95%CI 8:53-10:58) per report. Two months post-intervention, this was 6 minutes and 34 seconds (95% CI 6:25-6:43), reducing the time to process the remaining backlog (10,000 reports) by 54 days. Conclusion Applying LEAN techniques to HIV/AIDS surveillance in Portugal enabled delivery of crucial information to national and international HIV stakeholders through a 30% reduction in data processing time and optimization of data quality. Public health practitioners should consider LEAN techniques to improve data quality and efficiency of surveillance systems.
|
|---|